Investigation on electrical treeing behaviour at needle defects in cable insulation under AC voltage with the help of image processing algorithms and deep neural networks

In this paper, videos of voltage tests on needle defects in XLPE-MV-Cable-specimen are used to gain more information about the electrical treeing behavior in XLPE cable insulation. Accordingly, the maximum length of electrical treeing towards the opposite electrode inside the cable insulation will be investigated over time (treeing growth rate) to identify the influence of different parameters such as needle tip radius, gap distance and voltage level. Therefore, the growth rate of initiated electrical treeing (ET) during application of 50-Hz-AC has been investigated by using an image processing algorithm for image segmentation, edge detection and digital morphology. In contrast to conventional breakdown analyses, like time till treeing inception, inception-voltage or time till breakdown, this paper focuses on the treeing behavior over time. The automatic treeing identification will be evaluated manually, to check the performance of the program. This contribution shall state a first step into the automatic treeing identification by image processing algorithms and will give more insight in the treeing propagation in insulating material at an artificial needle tip failure.